Please fill out the following sections to the best of your ability, it will help us investigate bugs if we have this information at the outset. Screenshots are especially helpful, so please provide those if you can.
Platform : Website
Browser : Firefox
URLs: Jeff Davis Co. Search ( https://www.inaturalist.org/observations?nelat=31.77465403774715&nelng=-89.58658700633877&place_id=any&subview=map&swlat=31.39001008030971&swlng=-89.9762448340591&taxon_id=521318 );
Covington Co. Search (https://www.inaturalist.org/observations?place_id=2478&subview=map&taxon_id=521318)
Screenshots of what you are seeing :
Description of problem : Shapefile boundary for Jefferson Davis county is incorrect. When looking up occurrence records for Macrochelys temminckii, two identical observations are being captured by both the boundary of Jefferson Davis Co. and Covington Co… See screenshots, Jefferson Davis shapefile boundary is incorrect.
Step 1: Search Macrochelys temminckii in Jefferson Davis County, MS (there are no known records of this species in this county either)
Step 2: Search and compare the accurate results from Covington County, MS (this is where the actual records are from and the incorrect boundary of Jeff Davis overlaps into Covington.
Please use Filters->More Filters->Place instead of the Location search box.
I thought that the “Location” search box uses Locations in Google Maps database, while the “place” filter uses locations in iNaturalist’s database. If that were true, your issue would be with Google, not iNaturalist, but I can see it’s not that simple In this particular case, querying for “Covington County MS” in the Location box returns a place ID that is identical to the place ID used in the Place Filter search, but querying for “Jefferson Davis County MS” in the Location Box returns not a place ID, but a set of four coordinates that define a rectangular polygon that isn’t the county boundary. I haven’t seen that happen before. It’s like Google doesn’t know the county exists.
Thanks, I will use that next time. What Janet said is probably right - a Google maps issue. When downloading exported data from iNat the problem persists. I downloaded an export of all turtle records in Mississippi and the observations in question were included in the incorrect county there as well.
Yep that’s exactly what I was seeing, I tried a few other counties to check as well and those were all the same place ID that matched the iNat place filter boundary. Only Jefferson Davis for some reason.
for what it’s worth, i would not recommend that anyone look at that discussion, since it’s likely just going to cause unnecessary confusion. it looks like there’s a true issue described in this (Jefferson Davis, MS) thread related to the matching process between Google’s location results and iNaturalist’s places, though it’s easily worked around, as noted by jwidness. (and that method of searching for places is probably the better way to filter for geography in general on the Explore screen anyway.)
it may be worth noting that the problem described here appears to be different than what’s described in https://forum.inaturalist.org/t/searching-for-santa-rosa-ca-gives-wrong-bounding-box/22052. the problems described in that thread are technically working as designed, i think, but the design may not work well in these cases and may need to be rethought.
in general, i think the best way to redesign the location vs place filter is to reframe things as:
- a place filter that searches and returns only iNaturalist places
- a bounding box filter that allows you to either
- manually define a bounding box based on a set of coordinates (which could also work in concert with the Redo Search in Map button)
- define a bounding box based on a location from Google Maps.
- (define a bounding circle based on center point and km radius)
i think this would make it more clear how iNat places differ from bounding boxes and would eliminate the logic needed to match Google locations and iNaturalist places.